Celebrating in Your Success, Suffering in Your Defeat: Self-Conscious Emotion Understanding in Adolescents with Autism

Thursday, May 10, 2018: 11:30 AM-1:30 PM
Hall Grote Zaal (de Doelen ICC Rotterdam)
K. Jankowski and J. H. Pfeifer, University of Oregon, Eugene, OR

A two hit model of autism argues that adolescents with ASD experience a critical secondary wave of impairments in social cognition and peer interactions, enhanced by neural maturation and socioaffective changes related to puberty. Self-conscious emotions (SCEs), such as embarrassment and pride, are complex emotions representing self-awareness, perspective-taking, and social conceptual knowledge processing. SCEs are critical for supporting/maintaining interpersonal relationships; thus, atypical SCE understanding may underlie social difficulties in ASD.


We investigated the subjective experience and neural correlates of SCE understanding in adolescent males with high-functioning ASD and age-matched neurotypical (NT) males.


Study I investigated group differences in SCE attributions (the ability to recognize SCEs) and empathic SCEs (the ability/tendency to feel SCEs for others) in 56 adolescents (ASD=30, NT=26). It also explored associations with a triad of social cognitive abilities (self-awareness, perspective-taking, empathy) and autistic traits. Study II investigated the neural correlates of SCE processing in 52 adolescents (ASD=27, NT=25).

During an MRI scan, participants completed the novel Self-Conscious Emotions Task, which included 24 salient, ecologically-valid videos of adolescents in a singing competition. Videos represented two factors: emotion (embarrassment, pride) and perspective-taking (PT) demands (low, high). In low PT clips, singers’ emotions matched their performance (sing poorly, act embarrassed); in high PT clips, they did not (sing well, act embarrassed). Participants rated how intensely embarrassed and proud singers felt. They made congruent ratings, which matched the conveyed emotions (rating how embarrassed an embarrassed singer felt), and incongruent ratings, which did not match the conveyed emotions (rating how proud an embarrassed singer felt). Outside the scanner, participants rated how empathically embarrassed and proud they felt for the singers and completed social cognition tasks/questionnaires.

We conducted 2(group) x 2(emotion) x 2(PT) repeated measures ANOVAs investigating inferred and empathic SCE ratings. We also conducted 2x2x2 repeated measures ANOVAs investigating neural activity across the whole brain and in regions of interest.


Groups made similarly intense inferred congruent SCE ratings; however, the ASD group made more intense inferred incongruent SCE ratings during high PT demands. Inferred incongruent SCE ratings correlated negatively with PT abilities and positively with autistic traits. Groups made similarly intense empathic SCE ratings, which correlated positively with self-reported empathy.

Groups recruited broadly similar neural patterns during SCE processing. However, there were group differences within social cognition regions (medial prefrontal cortex, temporal pole), modulated by PT demands, and salience (anterior insula, anterior cingulate cortex) and sensorimotor regions, modulated by the situational context (singing quality).


Results reveal that adolescents with ASD can recognize SCEs and feel empathic SCEs. However, more intense incongruent SCE ratings during high PT demands suggest that emotion attributions may be more strongly impacted by the situational context. An over-reliance on contextual cues may reflect a strict adherence to rule-following and serve as a compensatory strategy for attenuated reflexive mentalizing. Additionally, adolescents with ASD broadly recruit similar neural patterns during SCE processing; however, atypicalities within social cognition, salience, and sensorimotor regions may reflect an over-reliance on abstract social conceptual knowledge. Implications for intervention are discussed.